Research map / Point clouds / Segmentation and detection
Annotator: A Generic Active Learning Baseline for LiDAR Semantic Segmentation
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Select voxel grids by VCD rather than uncertainty or random sampling | Annotator with MinkNet · Four LiDAR domain-transfer benchmarks, AL/ASFDA/ADA | mIoU: 53.7 vs 46.9; 54.1 vs 49.6; 57.7 vs 50.1 (SynLiDAR → KITTI, AL/ASFDA/ADA) | Helped |
| Select voxel grids by VCD rather than uncertainty or random sampling | Annotator with MinkNet · SynLiDAR → POSS, AL/ASFDA/ADA | mIoU: 44.9 vs 42.8; 48.2 vs 45.5; 52.0 vs 50.1 | Helped |
| Select voxel grids by VCD rather than uncertainty or random sampling | Annotator with MinkNet · KITTI → nuScenes and nuScenes → KITTI, AL/ASFDA/ADA | mIoU: 70.4 vs 66.0; 72.4 vs 67.5; 75.9 vs 73.1; 76.8 vs 73.1; 75.3 vs 69.7; 81.8 vs 76.7 | Helped |
| Increase selection voxel size from Δ₂=0.05 to Δ₂=0.25 (ablation) | Annotator · SynLiDAR → POSS (MinkNet, AL) | mIoU: 39.6 → 44.9 | Helped |
| Increase selection voxel size from Δ₂=0.05 to Δ₂=0.25 (ablation) | Annotator · SynLiDAR → POSS (MinkNet, ASFDA) | mIoU: 40.0 → 48.2 | Helped |
| Increase selection voxel size from Δ₂=0.05 to Δ₂=0.25 (ablation) | Annotator · SynLiDAR → POSS (MinkNet, ADA) | mIoU: 44.5 → 52.0 | Helped |
| Use a larger selection voxel size, Δ₂=0.3 instead of Δ₂=0.25 (ablation) | Annotator · SynLiDAR → POSS (MinkNet, AL) | mIoU: 44.9 → 45.1 | No clear effect |
| Use a larger selection voxel size, Δ₂=0.3 instead of Δ₂=0.25 (ablation) | Annotator · SynLiDAR → POSS (MinkNet, ASFDA) | mIoU: 48.2 → 48.3 | No clear effect |
Closest papers by meaning
- Point-to-Voxel Knowledge Distillation for LiDAR Semantic Segmentation (CVPR 2022 · similarity 0.91)
- 3D-AVS: LiDAR-based 3D Auto-Vocabulary Segmentation (CVPR 2025 · similarity 0.88)
- Spiral: Semantic-Aware Progressive LiDAR Scene Generation and Understanding (NeurIPS 2025 · similarity 0.88)
- Construct to Associate: Cooperative Context Learning for Domain Adaptive Point Cloud Segmentation (CVPR 2024 · similarity 0.88)
- Exploring Scene Affinity for Semi-Supervised LiDAR Semantic Segmentation (CVPR 2025 · similarity 0.88)
- TREND: Unsupervised 3D Representation Learning via Temporal Forecasting for LiDAR Perception (NeurIPS 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.